Exaggerated or persistent fear is a common occurrence in psychiatric disorders. Although much progress has been made in uncovering the neural basis of fear learning through studies of Pavlovian conditioning, little is known about how to reduce or eliminate traumatic memories in pathological cases, which is of great clinical concern. We have recently devised a novel behavioral paradigm that attenuates and prevents the return of fear memories. Our results indicate that subtle modifications to a commonly employed clinical treatment (exposure therapy) could greatly improve outcome, and reduce the potential for relapse in individuals suffering from post-traumatic stress disorder or specific phobias. In parallel, a new computational model developed in our collaborator's lab (Dr. Yael Niv, at Princeton University) provides a principled explanation for our results, and will help generate testable predictions to refine our protocol, and better understand the mechanisms that underlie the persistent updating of fear memories. In the proposed studies, we will first refine our protocol using computationally-derived predictions. Next, we will test the limits of this optimized protocol, and examine whether it may be used to target stronger, older, and more complex fear memories, with the hopes that it may eventually be employed to attenuate traumatic memories in clinical populations.